[Intel-wired-lan] [PATCH next-queue v2 3/3] igc: Add support for PTP getcrosststamp()

Vinicius Costa Gomes vinicius.gomes at intel.com
Fri Nov 20 19:07:21 UTC 2020


Hi Jakub,

Jakub Kicinski <kuba at kernel.org> writes:

> On Tue, 17 Nov 2020 17:21:48 -0800 Vinicius Costa Gomes wrote:
>> > Also, what is the point of providing time measurements every 1
>> > millisecond?  
>> 
>> I sincerely have no idea. I had no power on how the hardware was
>> designed, and how PTM was implemented in HW.
>
> Is the PTMed latency not dependent on how busy the bus is?
> That'd make 1ms more reasonable.

At least from the values of the registers I couldn't see any difference
if I was fully using the 2.5G ethernet link or not.


Cheers,
-- 
Vinicius


More information about the Intel-wired-lan mailing list